Calibration of Inertial Measurement Unit (IMU) Using Neural Network

摘要

In this paper, A new calibration method based Radial-Basis Function Artificial Neural Network (RBF) is proposed. A RBF neural network is designed to approach the error model of IMU and the outputs of IMU is compensated by this RBF neural network which realize the calibration and the compensation of IMU. The table experiment is carried out. The new method is compared with the classic calibration method, and the results indicate that this new calibration method can improve the accuracy of IMU more validly than the classic one.